Supporting Information for:

Transcription

1 Supporting Information for: Reticular Synthesis of Microporous and Mesoporous 2D Covalent-Organic Frameworks Adrien P. Côté,* Hani M. El-Kaderi, Hiroyasu Furukawa, Joseph R. Hunt, Omar M. Yaghi* Center for Reticular Chemistry, Department of Chemistry and Biochemistry, University of California, Los Angeles, 607 East Charles E. Young Drive, Los Angeles, CA, 90095, USA. Section 1 General Methods, Materials, and Characterization of Building Blocks and COFs p. S1 Section 2 Specific Synthetic Procedures p. S4 Section 3 Structure Modeling and Atomic Coordinates of COFs p. S6 Section 4 Section 5 Section 6 Solid-State 11 B MQ/MAS and 13 C CP/MAS Nuclear Magnetic Resonance Studies for COF-8, COF-10, and COF-6. Comparison of COF PXRD Patterns to Those Calculated From bnn and gra Models Scanning Electron Microscopy Imaging (SEM) of COF-8, COF- 10, and COF-6. p. S11 p. S28 p. S31 Section 7 Thermogravimetric Analysis. p. S35 Section 8 Non-Local Density Pore Size Distributions. p. S36 S1

2 Section S1: Methods, Materials, and Characterization of Compounds. All reactions were performed under nitrogen using either glovebox or Schlenk line techniques. Tetrahydrofuran was distilled from benzophenone sodium ketal. Anhydrous 1,4-dioxane (99.8%) and Acetone (99.8%, extra dry) was purchased from Acros Chemicals. Mesitylene (98%) was purchased from Fluka and was not dried prior to use. 2,3,6,7,10,11-hexahydroxytriphenylene (HHTP) was purchased from TCI. Boronic acids, 1,3,5-tris[(4-phenylboronic acid]benzene (BTPB), 4,4 -biphenyldiboronic acid (BPBA), 1,3,5-benzenetriboronic acid (BTBA) were prepared according to published procedures. 1 Elemental microanalyses of all products were performed at the University of California, Los Angeles, Department of Chemistry and Biochemistry. Fourier transform infrared (FT-IR) spectra ( cm -1 ) were obtained from KBr pellets using a Nicolet FT-IR Impact 400 system. Powder X-ray data were collected using a Bruker D8-Discover θ-2θ diffractometer in reflectance Bragg-Brentano geometry employing Ni filtered Cu Kα line focused radiation at 1600 W (40 kv, 40 ma) power and equipped with a Vantec Line detector. Radiation was focused using parallel focusing Gobel mirrors. The system was also outfitted with an anti-scattering shield which prevents incident diffuse radiation from hitting the detector, preventing the normally observed large background at 2θ < 3º. Samples were mounted on zero background sample holders by dropping powders from a wide-blade spatula and then leveling the sample surface with a razor blade. Given that the particle size of the as synthesized samples were already found to be quite monodisperse no sample grinding or sieving was used prior to analysis, we note, however, that S2

3 the micron sized crystallites lead to peak broadening. The best counting statistics were achieved by collecting samples using a θ step scan from with an exposure time of 10 s per step. No peaks could be resolved from the baseline for 2θ > 35 therefore this region was not considered for further analysis. S3

4 Section S2: Specific Synthetic Procedures Synthesis of COF-8. A 60 ml vial was charged with BTPB (0.50 g, 1.14 mmol), 2,3,6,7,10,11-hexahydroxytriphenylene (HHTP) (0.37 g, 1.14 mmol), and 50 ml of a 1:1 v:v solution of mesitylene:dioxane. The resulting suspension was sonicated for 30 minutes at room temperature then placed at 85 C for 3 days to afford a gray powder. The resulting powder was filtered off and was washed with dry acetone (3x20 ml). The resulting powder then was activated with acetone (2x50 ml) for 2 days then dried at 85 C/10-5 torr to afford COF-8 as gray powder (0.53 g, 71 %); IR (KBr, cm -1 ) (br), 1611 (vs), 1443 (w), 1352 (vs), 1245 (m), 1163 (w), 1072 (w), 1021 (m), 831 (m), 746 (m), 690 (w), 654 (w). Anal. Calcd for C 14 H 7 BO 2 : C, 77.13; H, Found: C, 76.39; H, Synthesis of COF-10: A 60 ml vial was charged with BPBA (0.50 g, 2.05 mmol), HHTP (0.44 g, 1.37 mmol), and 50 ml of a 1:1 v:v solution of mesitylene:dioxane. The reaction mixture was sonicated for 30 minutes then heated at 85 C for 3 days to afford a gray powder. The resulting powder was filtered off and was washed with dry acetone (3x20 ml). The powder was then activated with acetone (2x50 ml) for 2 days then dried at 85 C/10-5 torr for 12 h to afford COF-10 as a gray powder (0.53 g, 67% ); IR (KBr, cm -1 ): 3420 (br), 1607 (vs), 1536 (w), 1450 (m), 1444 (m), 1352 (vs), 1245 (s), 1168 (m), 1067 (m), 1021 (w), 863 (w), 822 (w), 736 (w), 965 (w). Anal. Calcd for C 12 H 6 BO 2 : C, 74.68; H, Found: C, 75.08; H, 3.39 S4

5 Synthesis of COF-6. A 60 ml vial was charged with BTBA (0.50 mg, 1.2 mmol), HHTP (0.39 g, 1.20 mmol), and 50 ml of a 1:1 v:v solution of mesitylene:dioxane. The reaction mixture was sonicated for 30 minutes at room temperature then heated at 85 C for 5 days to afford a gray powder. The resulting powder was filtered off and was washed with dry acetone (3x20 ml). The powder was then activated with acetone (2x50 ml) for 3 days then dried at 85 C/10-5 torr for 12 h to afford COF-6 as a gray powder (0.58 g, 76%); IR (KBr, cm -1 ) 3425 (br), 1596 (m), 1536 (w), 1495 (m), 1443 (m), 1311 (vs), 1245 (s), 1164 (s), 980 (w), 863 (m), 832 (w), 710 (w), 644 (w). Anal. Calcd for C 7 H 9 BO 2 : C, 67.70; H, Found: C, 66.1; H, References 1. For leading references to boronic acid synthesis see: (a) Morgan, A. B.; Jurs, J. L.; Tour, J. M. J. Appl. Polym. Sci. 2000, 76, (b) Goldschmid, H. R.; Musgrave, O. C. J. Chem. Soc. C 1970, S5

6 Section S3: Structure Modeling and Atomic Coordinates of COFs. Below are listed the crystallographic parameters, including atomic positions for COFs- 8,10, and -6. All models were generated using the Cerius 2 chemical structure-modeling software suite employing the crystal building module. Positions of atoms in the respective unit cells are listed as fractional coordinates in Tables S1-S3. Simulated PXRD patterns were calculated from these coordinates using the PowderCell program [W. Kraus, G. Nolze, J. Appl. Cryst. 29, 301 (1996)]. This software accounts for both the positions and types of atoms in the structures and outputs correlated PXRD patterns whose line intensities reflect the atom types and positions in the unit cells. Models were build adhering to the respective space group symmetries of bnn (P6/mmm) and gra (P6 3 /mmc) fitting trigonal HHTP SBUs to the respective 3-fold symmetric sites of each structure. S6

7 Table S1: COF8 Hexagonal P-6m2 a = b = , c = Å atom x y z B O C C C C C C C C C H H H S7

8 Table S2: COF10 Hexagonal P6/mmm a = b = , c = Å atom x y z B O C C C C C C C H H H S8

9 Table S3: COF12 Hexagonal, P-6m2 a = b = , c = Å atom x y z B O C C C C C H H S9

10 Refined Unit Cell Parameters Following Le Bail Extraction: Note the close correspondence between calculated and refined unit cell parameters. Table S4: COF a = b (Å) b (Å) Hexagonal COF-8 (P-6m2) (3) 3.630(4) COF-10 (P6/mmm) (3) 3.526(1) COF-6 (P-6m2) (3) (9) S10

11 Section S4: Solid-State 11 B MQ/MAS and 13 C CP/MAS Nuclear Magnetic Resonance Studies for COF-8, COF-10, and COF-6. High resolution solid-state nuclear magnetic resonance (NMR) spectra were recorded at ambient temperature on a Bruker DSX-300 spectrometer using a standard Bruker magic angle spinning (MAS) probe with 4 mm (outside diameter) zirconia rotors. Cross-polarization with MAS (CP/MAS) was used to acquire 13 C data at MHz (1). The 1 H and 13 C ninety-degree pulse widths were both 4 µs. The CP contact time was 1.5 ms. High power two-pulse phase modulation (TPPM) 1 H decoupling was applied during data acquisition. The decoupling frequency corresponded to 72 khz. The MAS sample spinning rate was 10 khz. Recycle delays betweens scans varied between 10 and 30 s, depending upon the compound as determined by observing no apparent loss in the 13 C signal intensity from one scan to the next. The 13 C chemical shifts are given relative to tetramethylsilane as zero ppm, calibrated using the methine carbon signal of adamantane assigned to ppm as a secondary reference. Multiple quantum MAS (MQ/MAS) spectroscopy was used to acquire 11 B data at MHz. The 11 B solution-state ninety-degree pulse width was 2 µs. TPPM 1 H decoupling was applied during data acquisition. The decoupling frequency corresponded to 72 khz. The MAS spinning rate was 14.9 khz. A recycle delay of 3 s was used. The 11 B chemical shifts are given relative to BF 3 etherate as zero ppm, calibrated using aqueous boric acid at ph = 4.4 assigned to ppm as a secondary reference. The 11 B spectrum of COF-5 (Côté, A.P.; Benin, A.I.; Ockwig, N.W.; O Keeffe, M.; Matzger, A.J.; Yaghi, O.M. Science 2005, 310, 1166) was used as a standard against which successful synthesis of co-condensed COFs was achieved. Note that the high sensitivity of 11B NMR to the immediate bonding environment of B is born out in the S11

12 peak shape of the spectra. The near perfect match of the peak shapes of COFs-6, -8, and 10 to that of COF-5 confirms that the expected C 3 O 2 B 1 rings in the structures have formed. In comparison we have also provided the spectra for the building blocks, which are markedly different that those of the products, indicated that the expected condensation reactions have indeed occurred. S12

13 Figure S1. Solid-state 11 B NMR spectrum for Model Compound COF S13

14 Figure S2. Solid-state 13 C NMR spectrum for Model Compound COF-5. 13C-COF S14

15 Figure S3. Solid-state 11 B NMR spectrum for 1,3,5-benzene-tris(4-phenylboronic acid) (BTBA). S15

16 Figure S4. Solid-state 13 C NMR spectrum for 1,3,5-benzene-tris(4-phenylboronic acid) (BTBA). HO B OH Carbon Chemical Shift (ppm) HO B OH B OH OH S16

17 Figure S5. Solid-state 11 B NMR spectrum for COF-8. The peak shape matches that of the model compound COF-5 and is very different from the starting material spectra. S17

18 Figure S6. Solid-state 13 C NMR spectrum for COF-8. The peak positions match those that are expected and the spectra is unchanged except for the new peaks corresponding to the HHTP moiety. S18

19 Carbon Chemical Shift (ppm) B O O S19

20 Figure S7. Solid-state 11 B NMR spectrum for 4,4 -biphenyl diboronic acid (BPBA). S20

21 Figure S8. Solid-state 13 C NMR spectrum for 4,4 -biphenyl diboronic acid (BPBA). Carbon Chemical Shift (ppm) HO HO B B OH OH S21

22 Figure S9. Solid-state 11 B NMR spectrum for COF-10. The peak shape and position match that of the model compound and are very different from the starting material. S22

23 Figure S10. Solid-state 13 C NMR spectrum for COF-10. The peak positions match those that are expected and the spectra is unchanged except for the new peaks corresponding to the HHTP moiety. Carbon Chemical Shift (ppm) B B O O S23

24 Figure S11. Solid-state 11 B NMR spectrum for 1,3,5-benzene triboronic acid (BTBA). S24

25 Figure S12. Solid-state 13 C NMR spectrum for 1,3,5-benzene triboronic acid (BTBA). The expected peaks are present and spinning side bands are present. HO B OH Carbon Chemical Shift (ppm) HO B 1 2 B OH OH OH S25

26 Figure S13. Solid-state 11 B NMR spectrum for COF-6. The peak shape matches that of the model compound and is different from the starting material. S26

27 Figure S14. Solid-state 13 C NMR spectrum for COF-6. The peak positions match those that are expected and the spectra is unchanged except for the new peaks corresponding to the HHTP moiety. Carbon Chemical Shift (ppm) B B O O B S27

28 Section S5: Comparison of COF PXRD Patterns to Those Calculated From bnn and gra Models Figure S15: PXRD pattern of COF-8 (top) compared to patterns calculated from Cerius 2 with the stacking of the layers in AA eclipsed stacking arrangement with P-6m2 space group (middle) and AB staggered (graphite) arrangement with P-6m2 space group symmetry (bottom). Note the pattern from the staggered model does not match the pattern of COF-8. COF-8 COF-8 Eclipsed Model Staggered Model Theta S28

29 Figure S16: PXRD pattern of COF-10 (top) compared to patterns calculated from Cerius 2 with the stacking of the layers in AA eclipsed stacking arrangement with P /6mmm space group (middle) and AB staggered (graphite) arrangement with P6/mmc space group symmetry (bottom). Note the pattern from the staggered model does not match the pattern of COF-10. COF-10 COF-10 Eclipsed Model Staggered Model Theta S29

30 Figure S17: PXRD pattern of COF-6 (top) compared to patterns calculated from Cerius 2 with the stacking of the layers in AA eclipsed stacking arrangement with P-6m2 space group (middle) and AB staggered arrangement with P6 3 /mmc space group symmetry (bottom). Note the pattern from the staggered model does not match the pattern of COF- 6. COF-12 COF-12 Eclipsed Model Staggered Model Theta S30

31 Section S6: Scanning Electron Microscopy Imaging (SEM) of COF-8, COF-10, and COF-6. Samples of all 2D COFs were prepared by dispersing the material onto a sticky carbon surface attached to a flat aluminum sample holder. The samples were then gold coated using a Hummer 6.2 Sputter at 60 mtorr of pressure in an argon atmosphere for 45 seconds while maintaining 15 ma of current. Samples were analyzed on a JOEL JSM-6700 Scanning Electron Microscope using both the SEI and LEI detectors with accelerating voltages ranging from 1kV to 15kV. S31

32 Figure S18: SEM image of COF-8 revealing platelet morphology. S32

33 Figure S19: SEM image of COF-10 revealing platelet morphology. S33

34 Figure S20: SEM image of COF-6 revealing platelet morphology. S34

35 Section S7: Thermogravimetric Analysis. Samples were run on a TA Instruments Q-500 series thermal gravimetric analyzer with samples held in platinum pans under atmosphere of nitrogen. A 5 K/min ramp rate was used. Figure S21: TGA trace for an activated sample of COF-8, COF-10, and COF-6. S35

36 Section S8: Non-Local Density Pore Size Distributions. The Pore Size Distribution of both compounds was calculated from these adsorption isotherms by the Non-Local Density Functional Theory (NLDFT) method using a cylindrical pore model [K. Schumacher, P. I. Ravikovitch, A. Du Chesne, A. V. Neimark and Klaus K. Unger, Langmuir, 22, 4648 (2000)]. Figure S22: Figure S22: NLDFT Pore Size Distribution and Fit of NLDFT Model to 700 Isotherm Data for COF Isotherm Data DFT Fit 200 Fitting Error = 0.4 % P/Po 18.7 Å Pore Width (Å) S36

37 S37

38 Figure S22: NLDFT Pore Size Distribution and Fit of NLDFT Model to Isotherm Data for COF Å Isotherm Data DFT Fit Fitting Error = 0.3 % P/Po Pore Width (Å) S38

39 Figure S23: NLDFT Pore Size Distribution and Fit of NLDFT Model to Isotherm Data for COF Isotherm Data DFT Fit Fitting Error = 1.1 % P/Po Pore Width (Å) S39